Although environmental factors play a major role in the etiology of young adult Hodgkin lymphoma (HL), genetic factors are also important. Originally healthy identical twins of young adult HL patients have a 99-fold increased risk of disease compared to no excess among fraternal twins of patients. We hypothesized that heritable factors were related to regulation of cytokine secretion because untreated HL patients have high circulating levels of certain cytokines (Th2 type) and depressed cellular immunity (a marker for Thl type cytokines). We previously compared cytokine levels in 50 unaffected identical twins of young adult HL cases (surrogate cases) to those of matched controls. Unaffected twins had statistically significantly higher levels of IL-6 and IL-8, but lower levels of IL-12, compared to controls. We then found that case-twins were less likely to carry the low-secreting IL-6 allele compared to controls (OR= 0.29, 95% CI = 0.1-0.87) (Cozen, 2004). We also found that cases were more likely to carry the low-secreting IL-12 allele. We propose to confirm these results and evaluate the role of other cytokine genotypes (IL-5, IL-9, IL-10, IL-13, IL-18, TNF-alpha) in young adult HL etiology by expanding the original study and recruiting more twin pairs and controls (to add 66 additional twin pairs plus a control for a total of 154 trios of case-twin, unaffected twin and control). We will use Illumina to identify 384 known functional and additional SNPs in the proposed cytokine genes in order to obtain maximum information and avoid missing unknown regulatory sites. Haplotypes will be developed from the raw SNP data and risk by single SNPs and haplotypes will be assessed in matched pairs (case-twin vs. control, and unaffected twin vs. control) using multivariate conditional logistic regression. The unaffected identical twin serves as a quality control for the genotype of the case-twin, and in situations where the case-twin is deceased, the identical twins' genotype can be substituted. With this sample size we will have 80% to detect an odds ratio of 1.6 and 1.8 if the minor allele frequency is 30% and 15%, respectively (the range of most of the known cytokine minor alleles). This study is important since it may will lead to identification of causal genetic factors which may offer alternative therapeutic strategies or prevention.